Evaluation of Floods and Landslides Triggered by a Meteorological Catastrophe (Ordu, Turkey, August 2018) Using Optical and Radar Data

Abstract

This study explores the potential of photogrammetric datasets and remote sensing methods for the assessment of a meteorological catastrophe that occurred in Ordu, Turkey in August 2018. During the event, flash floods and several landslides caused losses of lives, evacuation of people from their homes, collapses of infrastructure and large constructions, destruction of agricultural fields, and many other economic losses. The meteorological conditions before and during the flood were analyzed here and compared with long-term records. The flood extent and the landslide susceptibility were investigated by using multisensor and multitemporal data. Sentinel-1 SAR (Synthetic Aperture Radar), Sentinel-2 optical data, and aerial photogrammetric datasets were employed for the investigations using machine learning techniques. The changes were assessed both at a local and regional level and evaluated together with available damage reports. The analysis of the rainfall data recorded during the two weeks before the floods and landslides in heavily affected regions shows that the rainfall continued for consecutive hours with an amount of up to 68 mm/hour. The regional level classification results obtained from Sentinel-1 and Sentinel-2 data by using the random forest (RF) method exhibit 97% accuracy for the flood class. The landslide susceptibility prediction performance from aerial photogrammetric datasets was 92% represented by the Area Under Curve (AUC) value provided by the RF method. The results presented here show that considering the occurrence frequency and immense damages after such events, the use of novel remote sensing technologies and spatial analysis methods is unavoidable for disaster mitigation efforts and for the monitoring of environmental effects. Although the increasing number of earth observation satellites complemented with airborne imaging sensors is capable of ensuring data collection requirement with diverse spectral, spatial, and temporal resolutions, further studies are required to automate the data processing, efficient information extraction, and data fusion and also to increase the accuracy of the results.

A comparison among fuzzy multi-criteria decision making, bivariate, multivariate and machine learning models in landslide susceptibility mapping

“…Regarding the machine learning models, the random forest classifier (RFC) is based on a combination of decision tree classifiers and, therefore, it is considered a powerful supervised algorithm for solving binary classification tasks (Breiman 2001). The RFC model has been used in several LS studies, such as Chen et al (2018aChen et al ( , 2018b, Sevgen et al (2019), Nsengiyumva and Valentino (2020), Kocaman et al (2020) or Zhao et al (2020). Similarly to RFC model, Naïve Bayes classifier (NBC), a supervised probabilistic algorithm built on Bayes theorem, have been applied in several studies in recent years (Tsangaratos and Ilia 2016;He et al 2019;Chen et al 2020aChen et al , 2020bLee et al 2020;Lei et al 2020aLei et al , 2020b.…”

Large-Scale Landslide Susceptibility Mapping Using an Integrated Machine Learning Model: A Case Study in the Lvliang Mountains of China

“…The absolute value of the accuracy of the integrated model expressed by the ROC curve was not perfect, but it did improve the results compared with individual models. Such results supported the opinion that it is possible to combine different forecasts in an optimal prediction when multiple forecasts are available (Kocaman et al, 2020). This is mainly because the coupling model can combine the advantages of different models: The advantage of SVM is that high accuracy can be obtained on small sample training sets (Bui et al, 2016).…”

“…Hence, it is highly encouraged to produce "optimal" susceptibility models by combining multiple models (Reichenbach et al, 2018). However, it is of difficulties to determine how to best integrate multiple forecasts to obtain better results Rossi et al (2010), and limited attempts have been made on this issue, especially regarding the integration of machine learning models (Sevgen et al, 2019;Kocaman et al, 2020).…”

Detection of areas prone to flood risk using state-of-the-art machine learning models

“…Natural hazards cause extensive human and financial losses to humans worldwide each year Alam et al 2020Alam et al , 2021. In many parts of the globe, the acceleration of processes such as climate change, excessive surface waterproofing, and uncontrolled deforestation has led to an exponential increase in the severity and the number of natural hazards like drought and floods (Didovets et al 2019;Feng et al 2020;Kocaman et al 2020;Peptenatu et al 2020). Among the mentioned phenomena, the global climate changes are responsible for the amplification of the intensity and frequency of the torrential rains which are the main factor that generates the flood phenomenon (Shahid et al 2016;Markus et al 2018;Sepehri et al 2020).…”